We identified
pvl-1(
ga80) X in a screen for mutations causing a protruding vulva or Pvl phenotype (WBG 13.1,
p69). In
pvl-1(
ga80) hermaphrodites, one or more of P5.p, P6.p or P7.p remains undivided or adopts a 3! fate, leading to the generation of fewer vulval cells than normal, and a Pvl or Egl phenotype. In addition, P3.p, P4.p and P8.p often remain undivided rather than adopting the 3! fate.
pvl-1 hermaphrodites show a highly penetrant transformation of P12.p to a P11.p fate, similar to that seen for mutations in components of the Ras signaling pathway,
lin-44 wingless , and
mig-5 disheveled (Guo and Hedgecock, 1995 Mtg Abs,
p245). In addition,
pvl-1 animals are slightly uncoordinated and males mate poorly. A second allele of this gene,
sy324, was kindly provided by Andy Golden. Gene dosage experiments using a deficiency suggest that neither of these mutations is a null allele. We used transformation rescue experiments to clone
pvl-1. We identified overlapping cosmids, T21H4 and F43E12, that rescue the
pvl-1 mutant phenotypes, and then narrowed down the rescuing activity to a DNA fragment predicted to contain a single gene. Comparison of the sequence of full-length and partial cDNAs with genomic sequence from the Genome Sequencing Project showed that this gene encodes a predicted 812 amino acid protein with 30% identity to members of the armadillo/b-catenin class of adhesion/signal transduction molecules. armadillo/b-catenin proteins have two well-characterized functions. First, these proteins are localized to adherens junctions where they play a role in cell-cell adhesion through interactions with E-cadherins and a-catenin. Second, armadillo/b-catenin proteins function in wingless signaling pathways. In Drosophila, wg signal leads to the stabilization of arm protein and its accumulation in cytoplasm, where it functions in signal transduction. Recent experiments on vertebrate b-catenins have suggested that they can interact with transcription factors, and may translocate into the nucleus as a consequence of signaling. We are currently doing experiments to try to determine if
pvl-1 encodes a true C. elegans homolog of armadillo/b-catenin, or is a novel, armadillo -repeat-containing protein. We have built a fusion gene that encodes residues 1-763 of PVL-1 fused to GFP, and that rescues the
pvl-1 mutant phenotypes. Preliminary results show expression of this fusion protein in a number of tissues, including the Pn.p cells before induction, and
hyp7 and other hypodermal cells in larvae and adults. The GFP fusion protein appears to be cytoplasmically-localized in these tissues. We soon hope to verify the localization and expression pattern of Pvl-1 by staining with anti-Pvl-1 antibodies. Genetic epistasis experiments with
pvl-1(
ga80) have shown that the
ga80 mutation can suppress the ectopic inductions at P3.p, P4.p and P8.p caused by gain-of-function mutations in
let-60 ras and
mpk-1 MAP Kinase. However,
ga80 has much less of an effect on such inductions caused by loss-of-function mutations in
lin-1 and
lin-31, which encode transcription factors acting downstream of the ras pathway. One interpretation of these results is that
pvl-1 functions downstream of
let-60 and
mpk-1 during vulval induction, but upstream or parallel to
lin-1 and
lin-31. This would represent a novel function for an armadillo/b-catenin-related protein, and we are interested in determining if
pvl-1 is a target of the Ras signaling pathway in the vulva. For example, we would like to know if the amount or subcellular localization of PVL-1 changes upon Ras signaling. Alternatively,
pvl-1 might function in a novel signaling pathway that interacts with the Ras pathway during vulval development, perhaps a wingless pathway. We are testing if
pvl-1 interacts with wingless pathway genes such as
lin-44,
lin-17 frizzled (Sawa and Horvitz, 1995 Mtg Abs,
p455) or
mig-5 disheveled during vulval induction. In addition, we are doing yeast two hybrid experiements to identify proteins that interact with PVL-1.